Decline of Lake Michigan-Huron Levels Caused by Erosion of the St. Clair River

Transcription

1 Decline of Lake Michigan-Huron Levels Caused by Erosion of the St. Clair River W.F. & Associates Coastal Engineers (in association with Frank Quinn) April 13, 2005

2 Outline Problem Definition Understanding of Water Balance In Great Lakes Possible Causes of Head Decline Four Independent Analysis Conclusions and Future Studies

3 Relevant Water Level Gauges in the Study Area

4 Head Decline Between Lakes Michigan-Huron and Erie 3.5 Lake Michigan-Huron - Lake Erie Lake Michigan-Huron - Lake St.Clair Level Difference (m) Similar Trend Lake St.Clair - Lake Erie No change in Level Difference Year

5 Historic Regime Change (IJC, 1987) Regime Change Date Estimated Effect on Lake Huron Water Level (m) 6.1 m Navigation Channel Dredging 1855 to to Removal of Shoal from St. Clair Flats Sinking of Steamers Fontana and Martin Sand and Gravel Mining 1908 to Dredging 7.6 m Navigation Channel 1930 to Dredging 8.2 m Navigation Channel 1960 to NET EFFECT 1855 to to 0.46

6 In the Recent 30+ Years (1971 present) No significant known human actions influencing Lake Michigan-Huron and the St. Clair River Direct impact of navigation channel deepening project would have ceased to be a factor after 10 years

7 What Caused this Head Decline? Post-glacial rebound impacts Net basin supply change and shift Erosion of the St. Clair River???

8 Outline Problem Definition Understanding of Water Balance in the Great Lakes Possible Causes of Head Decline Four Independent Analyses Conclusions and Future Studies

9 Possible Causes of Head Decline Post-glacial rebound impacts Net basin supply change and shift Erosion/dredging in the St. Clair River

10 Differential Rebound Basin tilting effect min. glacial rebound 3 cm rise in last 30 years

11 Impact in Last 30 Years 3 cm lake level rise on Lake Erie 1.5 cm lake level rise on Lake MH Conclude: no significant contribution (1.5 cm) on head drop between Lake MH and Erie

12 Possible Causes of Head Decline Post-glacial rebound impacts Net basin supply change and shift Erosion/dredging in the St. Clair River

13 Net Basin Supply (NBS) NBS is the total net water supply to a lake Two methods of calculating NBS Components Method (GLERL) Residuals Method (EC and USACE)

14

15 Relative NBS (Lake Erie/Lake Huron) NBS Ratio (Erie/Michigan-Huron) Year Components Method Residual Method 10 Year Moving Average 10 Year Moving Average Data using Components and Residual methods diverge.

16 Key Points on an NBS Shift Residual NBS shift is probably produced by the incorrect flow data If it is occurring (and this seems unlikely or at least unproven) NBS shift would not have a significant contribution to the observed head drop (almost certainly less than 4 cm)

17 Possible Causes of Head Decline Post-glacial rebound impacts Net basin supply change and shift Erosion/dredging in the St. Clair River

18 Outline Problem definition Understanding of water balance in the Great Lakes Possible causes of head decline Four independent analysis for erosion Conclusions and Future Studies

19 Four Independent Analyses Hydraulic analysis using gage relationships Normalization analysis using water balance equation on Lake Erie GIS analysis on historical bathymetry change Numerical modeling

20 Historical Change of Relationship between Heads (MH E) and Lake Level (Cleveland) Head Difference in Lake Level Between Lakes Huron and Erie Before After 1960 Difference in Lake Level (m) y = x R 2 = y = x R 2 = y = x R 2 = Lake Level in Lake Huron (m)

21 Four Independent Analyses Hydraulic analysis using gage relationships Normalization analysis using water balance equation on Lake Erie GIS analysis on historical bathymetry change Numerical modeling

22 Residual Head (Recorded Head Normalized Head + 2.5) (1885) (1900) Head (m) (1948) (1999) Year

23 Compare to IJC Estimates (1985, 1987) Estimated Level Difference Decrease (IJC 1987) ' Navigation Channel Lakers Sunk Sand and Gravel 25' Navigation Mining Channel 27' Navigation Channel to -0.21m m -0.09m -0.05m -0.13m -0.23m IJC Estimated Head Decrease Caused by Erosion (, 2005) Head Decline (m) IJC Estimated plus Diversion Chicago Diversion Flow Increase (2.8 to 283 Chicago Diversion Limited to 90 m 3 /s 1.8 Welland Cannal Diversion (31 to 260 m 3 /s) Year Long Lac/Ogoki Diversion (159 m 3 /s)

24 Key Points from Normalization Analysis Reproduces well the historic regime change events and diversions Clearly indicates continuous head decline since 1971, in which the head variation caused by natural climatic change is filtered out The head decline must be caused mostly by regime change of the St. Clair River

25 Four Independent Analyses Hydraulic analysis using gage relationships Normalization analysis using water balance equation on Lake Erie GIS analysis of historical bathymetry change Numerical modeling

26 Historic Dredging and Events Regime Change Date Estimated Effect on Lake Huron Water Level (m) 6.1 m Navigation Channel Dredging 1855 to to Removal of Shoal from St. Clair Flats Sinking of Steamers Fontana and Martin Sand and Gravel Mining 1908 to Dredging 7.6 m Navigation Channel 1930 to Dredging 8.2 m Navigation Channel 1960 to NET EFFECT 1855 to to 0.46

27 Erosion of St. Clair River Channel 1867 Bathymetry 1929 Bathymetry 1971 Bathymetry (1948 in Figures) 2000 Bathymetry

28 1867 Bathymetry Upper St. Clair River

29 Comparison Bathymetry Net Erosion

30 Upper River Erosion and Accretion Patterns

31 Comparison Bathymetry Net Erosion

32 0 Profile Profile (Datum Unknown) ft Channel 27ft Channel Depth (m) profile Profile (IGLD 55, sloping) (IGLD55 Sloping) Profile (IGLD 85, sloping) 1929 Profile (US Standard Datum 1902, sloping) Correction for Vertical Datum has not yet been applied West side of Channel Distance along Profile (m) East side of Channel

33 Key Points From GIS Analysis of Bathymetry Change The St. Clair River has eroded significantly between 1971 and 2000 The erosion mostly explains the decline of Lake Michigan-Huron levels Continuous erosion results primarily in an upstream lake drop

34 Four Independent Analyses Hydraulic analysis using gage relationships Normalization analysis using water balance equation on Lake Erie GIS analysis on historical bathymetry change Numerical modeling

35 Numerical Modeling Two numerical models applied in the St. Clair River and parts of Lakes Huron and St. Clair RMA2 a 2D hydrodynamic model MISED a 3D hydrodynamic and sediment transport model

36 RMA2 Model Originally from USACE, Detroit District Model was developed and calibrated by USACE/USGS using ADCP data Model domain adjusted and included The St. Clair River Part of Lake Huron Part of Lake St. Clair River

37 Lake Huron Level Drops with 2000 Bathymetry (using the same flow mean flow) Water Surface Elevation Profile cm Drop 1948 Bathymetry 2000 Bathymetry With 1971 Bathymetry Water Surface Elevation (m) With 2000 Bathymetry Distance from Lakeport (km)

38 Compare to IJC Estimates (1985, 1987) Estimated Level Difference Decrease (IJC 1987) ' Navigation Channel Lakers Sunk Sand and Gravel 25' Navigation Mining Channel 27' Navigation Channel to -0.21m m -0.09m -0.05m -0.13m -0.23m IJC Estimated Head Decrease Caused by Erosion (, 2005) Head Decline (m) IJC Estimated plus Diversion Chicago Diversion Flow Increase (2.8 to 283 Chicago Diversion Limited to 90 m 3 /s 1.8 Welland Cannal Diversion (31 to 260 m 3 /s) Year Long Lac/Ogoki Diversion (159 m 3 /s)

39 MISED Modeling in-house model - a 3D hydrodynamic and sediment transport model Very detailed modeling application (2 to 4 meter grid resolution near the month of Black River) Model Domain Part of Lake Huron Upper part of the St. Clair River

40

41 Calibrated with USACE ADCP Data (X- Section 06)

42 Flow Profile Calibration at Point A on Cross-Section 17 ( , ) Flow Direction Profile at Point A on Cross-Section 17 ( , ) Computed Measured Measured Measured 2.00 Computed Measured Measured Measured Depth (m) 6.00 Depth (m) Flow Velocity (m/s) Flow Direction (Degree to East) Measured ADCP Data 0.00 Flow Velocity Profile at Point B on Cross-Section 17 ( , ) 0.00 Flow Direction Profile at Point B on Cross-Section 17 ( , ) 1.50 Run 5 Modeled Data Computed Measured Measured Measured Velocity (m/s) Depth (m) Computed Measured Measured Depth (m) Measured Flow Velocity (m/s) Flow Direction (Degree to East) Distance (m) X-Section 17

43 Erosion Potential Red Fine Gravel Yellow Very Fine Gravel Green Medium Sand Blue Finer than Medium Sand

44

45 Riverbed Erosion Erosion generally caused by: More sediment moving out of an area than into that area Exposure of an irreversibly erodible sediment (removal of lag) The possible causes for recent erosion include: aggregate mining coastal shore protection riverbank protection, and indirectly, dredging Ship-enhanced erosion and transport

46 Outline Problem Definition Understanding of Water Balance in the Great Lakes Possible Causes of Head Decline Four Independent Analyses Conclusions and Future Studies

47 Conclusions What has happened Water level data shows a previously undetected/unexplained 25 to 35 cm drop in head difference between M-H and E over the past 30 to 35 years High lake levels between 1970 and 1998 had previously masked the full extent of the head drop

48 Conclusions What has caused it In the last 30 to 35 years, of the 25 to 35 cm drop: Glacial rebound accounts for less than 2 cm (Erie rise) NBS shift accounts for less than 4 cm (even this unlikely) Continuous erosion may have raised Lake Erie by 2 cm No significant contribution from change due to diversions Numerical model of 1971 and 2000 bathymetry (representing significant erosion) can account for 23 cm of change all resulting from a fall of Lake Michigan-Huron due to increased flow capacity on the St. Clair River

49 Conclusions What about the future Both the hydraulic analysis and the normalization analysis suggest the decline in head difference (due mostly to a fall of Lake Michigan) is ongoing Lake level change due to erosion is irreversible Long-term cycles suggest falling lake levels over the next 80 years Latest climate change also predicts reduction in lake levels

50 Conclusions What triggered and sustains the erosion Lakes Michigan-Huron have been relatively stable for 2000 to 3000 years due to a stable outlet Recent changes that may have contributed to triggering and sustaining erosion: Sand mining Dredging (indirectly) Coastal protection and structures River bank protection Ship-enhanced erosion and transport

51 Future Studies Bathymetry survey of the upper river and lake (and review any 1980s, 1990s bathymetry) Boreholes of the eroding area, ROV, geophysical surveys 3D modeling of waves, currents, sand transport, cohesive sediment erosion and morphodynamics Geomorphic assessment, detailed sediment budget Explanation of the erosion Development/testing of solutions

52 Outline Pdf s of erosion Physical model of St. Clair and Black River Detroit River

53 Acknowledgements & Associates completed this work through funding by the GBA Foundation Cooperation and assistance of the USACE, Environment Canada and the Great Lakes Environmental Research Laboratory is gratefully acknowledged Many thanks to Frank Quinn for his valuable input and review of our work.

54 Detroit River, Fighting Island to Belle Isle - Analysis of River Bed Erosion W.F. & Associates Coastal Engineers April 13, 2005

55

56

57 Bed Change Between 1925 to 2000

58 Profile G - Detroit River 0 Profile G 2000 Data Profile G 1925 Data Depth (m) IGLD Right Bank (north) Distance along Profile (m) Left Bank (south)

59 Profile E - Detroit River 0 Profile E 2000 Data Profile E 1925 Data Depth (m) IGLD Right Bank (north) Distance along Profile (m) Left Bank (south)

60 Profile C - Detroit River 0 Profile C 2000 Data Profile C 1925 Data Depth (m) IGLD Right Bank (north) Distance along Profile (m) Left Bank (south)

61 Profile B - Detroit River 0 Profile B 2000 Data Profile B 1925 Data Depth (m) IGLD Right Bank (north) Distance along Profile (m) Left Bank (south)

62 Profile A - Detroit River 0 Profile A 2000 Data Profile A 1925 Data Depth (m) IGLD Right Bank (north) Distance along Profile (m) Left Bank (south)

63 Profile D - Detroit River 0 Profile D 2000 Data Profile D 1925 Data Depth (m) IGLD Right Bank (north) Distance along Profile (m) Left Bank (south)

64 Profile F - Detroit River 0 Profile F 2000 Data Profile F 1925 Data Depth (m) IGLD Right Bank (north) Distance along Profile (m) Left Bank (south)

65 Bed Changes Between 1925 and 2000 Generally, the river bed is in an erosional state About 1 metre of downcutting on average for this section of the river bed in 75 years

66 River Scour Assessment Water level data collection and analysis Bed sediment erodibility 100 year scour depth assessment River scour under natural river flows Storm surge impact Ship traffic, ice jam, and global warming impacts

67 Bed Sediment Erodibility for Hard Clay E erosion rate in mm/hr a constant (=0.06) E = at 1.5 T = (τ b - τ cr )/ τ cr τ b bed shear stress τ cr critical shear stress for erosion (=2.25 pa) Erosion Rate (mm/hr) Erosion Rate of Hard Clay (clear water) St. Joseph's Til ( dark,hard) JWK ('91) Port Stanley Till E=a*T^ Shear Stress (Pa)

68 100 Years of Erosion Under Natural Flow Estimated flow velocity is in range of 0.6 to 0.8 m/s Consistent with the flow velocity measured in the river In total 0.4 m erosion is predicted for pure natural river flow (monthly data) over 100 years according to the erodibility equation

69 Storm Surge Impact Strong winds are most likely during fall and early spring The setup and setdown of lake levels increases flow velocity in the river Flow velocity depends on surge strength and duration Use daily data to estimate surge impact from 1970 to 2003

70 Storm Surge Impact Estimated flow velocity: 0.2 to 1.0 m/s In total about 1.4 m erosion is predicted for storm surge plus natural flow (daily data) over 100 years (using 30 years data)

71 Ship Traffic Impact Ship propeller erosion Function of ship size/draft, propeller type, and traffic Largest vessel with full power can generate about 4 m/s flow velocity near bed (depth 13 m) at the project site However, vessel speed is limited to 10.2 knots (5 m/s) Ship traffic data required Additional local erosion (about 0.5 m) may be caused by ship traffic at some locations Ships have less impact in other areas

72 Global Warming The temperature in the Great Lakes region could rise 2 to 4 C by the end of the 21 st century Precipitation could increase by 25% More intense rainstorms Great Lakes levels are expected to fall by 1.5 to 8 feet (0.5 m to 2.4 m);

73 Global Warming Impact on River Scour Global Warming Impact Downcutting Rate (m/100 year) Lake Level Drop (m) The downcutting rate may be more than that predicted because of more storm surges caused by global warming

74 Verification Downcutting Measured and from bathymetry comparison 1.25 m (max. 2.0 m) Predicted 1.36 m Flow velocity m/s m/s

75 Predictions for the next 100 years Assume the flow condition to be unchanged Natural flows the same as past Storm Surge the same as past Global warming lake level drop of 0.5 m Ship traffic not considered 1.32 m of additional downcutting by 2103

76 Preliminary Findings Bathymetry Comparison River bed changes since 1925 year Most of the river was in downcutting state The average erosion in the reach from Fighting Island to Belle Isle Island is about 1 m

77 Preliminary Findings Dynamic Analysis The analysis predicted the observed erosion in the past 100 years Bed changes estimated over past 100 years About 0.4 m scour caused by natural flow Additional 1 m scour contributed by storm surges

78 Preliminary Findings Dynamic Analysis Surges are the main driving force for scour Ship traffic contributes to river scour, particularly at the project site More erosion will be expected due to global warming Ice jams likely do not have a significant influence on river scour